Research And Optimization For Multiband Spectrum Sensing In Cognitive Radio

The wireless spectrum is one of the most valuable and scarce natural resources. With the dramatic increase of service quality and channel capacity in wireless networks, the scarcity of spectrum resources has been a serious problem. Cognitive radio (CR) has been widely considered as one of the prominent solutions to tackle the spectrum scarcity. Spectrum sensing (SS) is a key function of CR to prevent the harmful interference with primary users (PUs) and identify the available spectrum for improving the spectrum’s utilization. While the majority of existing research has focused on single-band spectrum sensing (SBSS), multiband spectrum sensing (MBSS) represents great promises toward implementing efficient cognitive net works compared to SBSS. MBSS is expected to significantly enhance the network’s throughput and provide better channel maintenance by reducing handoff frequency. MBSS has recently caught the attention since it can significantly enhance the throughput of second user (SU). There are several scenarios where MBSS can be encountered. First, many modern communication systems and applications require a wideband access. The wideband spectrum can be divided into multiple subbands orsubchannels. Thus, the problem becomes a multiband detection problem. Secondly, when an SU wants to minimize the data interruptions due to the return of PUs to their bands, seamless handoff from one band to another becomes vital. Therefore, the SU must have backup channels besides those channels it is currently using. With MBSS, the SU does not only have a set of candidate channels, but it can also reduce handoff frequency. Thirdly, When an SU wants to achieve higher throughput or maintain a certain QoS, then it may transmit over a larger bandwidth, and this is primarily enabled by accessing multiple bands. Finally, In cooperative communications, multiple SUs may share their detection results among each other. However, if each SU monitors a subset of subchannels, and then shares its results with others, then the entire spectrum can be sensed, and consequently, more opportunities are explored for spectrum access. This dissertation focuses on the problem of MBSS in cognitive radio and the main contributions of this dissertation are: 1) This dissertation studies the mathematical model of multiband joint detection (MJD). The MJD problem can be formulated as a constrained optimization problem, whose goal is to maximum the aggregated opportunistic throughput of a CR system under some constraints on the interference to the PUs. An immune clone algorithm (ICA) is proposed to solve this problem. The performance of the proposed method is analyzed and compared with the genetic algorithm (GA) based technique through computer simulations. The experiment results show that the proposed method offers higher aggregate opportunistic rates under the same interference to PUs than GA and has more robust stability and efficiency.2) In MJD of wideband sensing, the most challenge is to set the optimal decision thresholds due to the non-convex nature of the problem. This dissertation proposes the Branch Reduce and Bound algorithm with Convex Relaxation (BRBCR) technique to optimize the problem which can be transformed into a Monotonic Optimization Problem (MOP). The performance of the proposed method is analyzed through computer simulations. Experiment results show that this method can significantly improve the system performance as compared with the conventional convex optimization method. The convergence speed of the proposed method is two orders of magnitude faster than the Polyblock Algorithm (PA) and the conventional Branch Reduce and Bound (BRB) algorithm. Even though the number of channels is16and the convergence precision is10-6, this method can converge within16seconds. In addition, the proposed algorithm can also provide an important benchmark for evaluating the performance of other heuristic algorithms targeting with the same problem.3) The conventional method of joint optimzation MBSS and resource allocation tends to ignore the system throughput obtained when SU coexists with the PU. However, there is a large loss of system throughput if ignoring the part of system throughput when PU has the ability to tolerate strong interference. In addition, the conventional iterative optimization method does not consider the coupling constraint and has the poor performance when the coupling constraint becomes much more stringent. A iterative optimization algorithm based on penalty framework and DC programming is proposed to overcome the above problems. Experiment results show that the proposed method can significantly improve the system performance as compared with the conventional iterative optimization algorithm.4) The Shannon capacity of the channel is always as the optimization goal in the conventional methods of joint optimzation MBSS and resource allocation which does not considering the complexity of the communication system. However, shannon capacity can only be considered as an upper limit of the channel capacity which is related to the BER and coding method in real systems. This dissertation proposes a iterative optimization algorithm based on penalty framework and DC programming to solve the problem of joint optimization MBSS and bit loading while guaranteeing a target average bit error rate (BER). Experiment results show that the proposed method can significantly improve the system performance as compared with the bit loading method.5) Much of the present research in the area of spectrum sensing scheduling (SSS) assumes that the occupancy of any particular frequency subband is independent of other subbands. However, in practice subband occupancy is likely to be correlated, for instance, due to the use of wideband transmission signals. In this dissertation, a method of SSS using the prior knowledge of such correlation of subbands is proposed. The method reduces the dimension of the problem firstly and then uses the ICA to solve it. Through analysis and numerical experiments, we demonstrate that the proposed method significantly outperforms the conventional method of SSS.In summary, this dissertation investigates several open problems and provides theoretical optimization model and solution methods to further exploit the performance of MBSS. Our works have their academic and practical value on promoting the advancement of the researches and applications in multiband cognitive radio networks.